Integrating Semantic Understanding and Textual Features for Fake News Detection Using Deep Learning
Bibliographic record
Abstract
The intensive technological change has triggered the introduction of significant changes in the channels of conveying and receiving information. Social media networks have become the leading ways of relaying information, that have the potential of reaching a large number of people within a short period of time. However, in the same breath, the same phenomenon has intensely boosted the spread of disinformation or what is popularly known as fake news, which has a reassuring impact on the sociocultural dynamics. The dissemination of fake news undermines the perceived integrity of sources of information, creating biased views and creating illusions about the relevant issues. Scholars and practitioners, in their turn, have increasingly applied the methodologies of artificial intelligence (AI) and machine learning (ML), to come up with more effective solutions to detecting fake news. ML, a subdivision of AI that is focused on algorithm conception that derives predictive functionality of data, proved to be significantly successful in addressing a variety of challenges, such as detecting non-true content. This paper presents an efficient approach to prediction of misinformation through a fused deep-learning system based on semantic-analysis. The theory behind the suggested methodology will combine the concepts of Artificial Neural Networks (ANNs) and Recurrent Neural Networks (RNNs) and will be aimed at analyzing the patterns of text and language arrangements, respectively. In addition, semantic analysis has also been integrated in order to increase the predictability by considering the semantic background and the contextual background of the textual substance being analyzed. The derived plan provides real-time tool of identifying false content, thus eliminating its negative impact on society.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".